Contrast enhancement in atherosclerosis development in a mouse model: in vivo results at 2 Tesla

To develop an MRI method for the evaluation of contrast enhancement in early atherosclerotic plaque development in the abdominal aorta of a mouse model. Male apoE^–/– mice from three groups, respectively 4 (n = 6), 8 (n = 11) and 16 (n = 4) weeks were included. Axial T1 spin echo images of the abdominal aorta were obtained above and below the renal arteries (90 m spatial resolution) before and over 1 h after the injection of a macromolecular contrast agent. Signal enhancement was measured in the vessel wall and compared to histological features. Maximal arterial wall signal enhancement was obtained from 16 to 32 min post injection. During this time, the signal-to-noise ratio increased by a factor up to 1.7 in 16 week mice and 2.7 and 2.4 in 8 and 4 weeks mice, respectively. The enhancement of the arterial wall appeared less pronounced in the oldest mice, 16 weeks old, exhibiting more advanced lesions. Using a macromolecular gadolinium agent, contrast uptake in atherogenesis varies with lesion stage and may be related to vessel-wall permeability. Dynamic contrast-enhanced MRI may be useful to evaluate the atherosclerotic plaque activity in mice.     

133Cs and 23Na MAS NMR Spectroscopy of Molybdate Crystallization in Model Nuclear Glasses

Sodium borosilicate base glasses modeled on French nuclear waste materials were prepared to test the dependence of crystallization product quantity and distribution on cesium‐ and molybdenum‐loading and glass cooling rate. Scanning electron microscopy shows the presence of micrometer‐sized domains of Mo‐rich crystalline precipitates. X‐ray diffraction patterns are complex but reveal the presence of sodium molybdates and CsNaMoO₄·2H₂O. ¹³³Cs and ²³Na magic‐angle spinning NMR spectroscopy exhibit distinct peaks for glassy and crystalline phases which can be quantified to obtain the identities of the individual compounds that are formed as well as the fractions of these nuclei in particular crystalline phases. In these model systems, 1 mol% Mo can be entirely incorporated into the glassy network whereas 2.5 and 5 mol% Mo produce significant quantities of crystalline phases, with little dependence on cooling rate. Cesium content appears to have a weak influence on crystallization behavior. Sodium molybdate and sodium‐cesium molybdate hydrate are the dominant devitrification phases in all cases.     

Testing of starch-based carbohydrate polymer coatings for enhanced urea performance

In this study, starch–urea–borate adhesives were developed for coating the slow release urea. The physical properties of the developed adhesives were studied as a function of temperature, heating time, stirring rate, and pH. It was found that for certain specific adhesive composition, pH and stirring rate, the complete gelatinization time and corresponding adhesive viscosity do not remain constant with temperature. The suspension heated at 75°C reached its maximum viscosity after 21 min of heating, thereafter, remained constant over time. In contrast, the suspension heated at 80°C reached its peak viscosity after 12 min of heating. Further heating after 12 min caused a steady decrease in viscosity from its peak value of 450–339 cP. Once the adhesive physical properties were completely understood, a dripping solution technique was used to coat the urea granules with coating thickness ranging from 0.15 to 0.7 mm. It was noticed that the overall nutrients release time of the coated urea was three times higher than the uncoated urea. It was also concluded that the mechanical strength of coated urea strongly depends on the adhesive composition and coating thickness.     

A New and Accurate Modeling Technique for Glass Resistance in Multiphase Multielectrode Glass Furnaces

One of the design elements involved in sizing the electrical power equipment for glass furnaces is the determination of the glass resistances between the operating electrodes. All methods used for this purpose to date are approximate and are based on a simple model considering the resistance between only two electrodes at the time. This paper presents a technique to develop a resistance model for any general configuration of electrodes and supply voltages in glass furnaces. The technique is based on relating the glass conductivity as represented by Ohm's law to the electrostatic property as represented by Gauss's law. The resistance model is then derived in a matrix form using the bus admittance frame of reference. A digital computer program has been developed to implement the proposed technique and example results are presented.     

Interpretation of way-finding healthcare symbols by a multicultural population: Navigation signage design for global health

The interpretation of way-finding symbols for healthcare facilities in a multicultural community was assessed in a cross-sectional study. One hundred participants recruited from Al Ain city in the United Arab Emirates were asked to interpret 28 healthcare symbols developed at Hablamos Juntos (such as vaccinations and laboratory) as well as 18 general-purpose symbols (such as elevators and restrooms). The mean age was 27.6 years (16–55 years) of whom 84 (84%) were females. Healthcare symbols were more difficult to comprehend than general-purpose signs. Symbols referring to abstract concepts were the most misinterpreted including oncology, diabetes education, outpatient clinic, interpretive services, pharmacy, internal medicine, registration, social services, obstetrics and gynecology, pediatrics and infectious diseases. Interpretation rates varied across cultural backgrounds and increased with higher education and younger age. Signage within healthcare facilities should be tested among older persons, those with limited literacy and across a wide range of cultures.     

Optimization of parallel firewalls filtering rules

As filtering policies are getting larger and more complex, packet filtering at firewalls needs to keep low delays. New firewall architectures are needed to enforce security and meet the increasing demand for high-speed networks. Two main architectures exist for parallelization, data-parallel and function-parallel firewalls. In the first, packets are distributed across a set of identical firewalls that implement the entire policy. In the second, each firewall implements a subset of the policy with a fewer number of rules, but the packets have to be duplicated and processed by all the firewalls. This paper proposes a new architecture function-parallel with pre-processing that combines the advantages of both architectures. The proposed architecture has the advantage of not duplicating the data, so that the processing time can be significantly reduced. Moreover, our architecture enables stateful inspection of packets, which is necessary to prevent multiple types of attacks. The performances of this architecture have been proven to be scalable for large security policies.

     

QRS detection based on wavelet coefficients

Electrocardiogram (ECG) signal processing and analysis provide crucial information about functional status of the heart. The QRS complex represents the most important component within the ECG signal. Its detection is the first step of all kinds of automatic feature extraction. QRS detector must be able to detect a large number of different QRS morphologies. This paper examines the use of wavelet detail coefficients for the accurate detection of different QRS morphologies in ECG. Our method is based on the power spectrum of QRS complexes in different energy levels since it differs from normal beats to abnormal ones. This property is used to discriminate between true beats (normal and abnormal) and false beats. Significant performance enhancement is observed when the proposed approach is tested with the MIT-BIH arrhythmia database (MITDB). The obtained results show a sensitivity of 99.64% and a positive predictivity of 99.82%.     

Unsupervised deep learning approach for network intrusion detection combining convolutional autoencoder and one-class SVM

Applied Intelligence - With the rapid advancement in network technologies, the need for cybersecurity has gained increasing momentum in recent years. As a primary defense mechanism, an intrusion...     

Top-Down Neural Attention by Excitation Backprop

We aim to model the top-down attention of a convolutional neural network (CNN) classifier for generating task-specific attention maps. Inspired by a top-down human visual attention model, we propose a new backpropagation scheme, called Excitation Backprop, to pass along top-down signals downwards in the network hierarchy via a probabilistic Winner-Take-All process. Furthermore, we introduce the concept of contrastive attention to make the top-down attention maps more discriminative. We show a theoretic connection between the proposed contrastive attention formulation and the Class Activation Map computation. Efficient implementation of Excitation Backprop for common neural network layers is also presented. In experiments, we visualize the evidence of a model’s classification decision by computing the proposed top-down attention maps. For quantitative evaluation, we report the accuracy of our method in weakly supervised localization tasks on the MS COCO, PASCAL VOC07 and ImageNet datasets. The usefulness of our method is further validated in the text-to-region association task. On the Flickr30k Entities dataset, we achieve promising performance in phrase localization by leveraging the top-down attention of a CNN model that has been trained on weakly labeled web images. Finally, we demonstrate applications of our method in model interpretation and data annotation assistance for facial expression analysis and medical imaging tasks.     

Recognizing arabic handwritten characters using deep learning and genetic algorithms

Automated techniques for Arabic content recognition are at a beginning period contrasted with their partners for the Latin and Chinese contents recognition. There is a bulk of handwritten Arabic archives available in libraries, data centers, historical centers, and workplaces. Digitization of these documents facilitates (1) to preserve and transfer the country’s history electronically, (2) to save the physical storage space, (3) to proper handling of the documents, and (4) to enhance the retrieval of information through the Internet and other mediums. Arabic handwritten character recognition (AHCR) systems face several challenges including the unlimited variations in human handwriting and the leakage of large and public databases. In the current study, the segmentation and recognition phases are addressed. The text segmentation challenges and a set of solutions for each challenge are presented. The convolutional neural network (CNN), deep learning approach, is used in the recognition phase. The usage of CNN leads to significant improvements across different machine learning classification algorithms. It facilitates the automatic feature extraction of images. 14 different native CNN architectures are proposed after a set of try-and-error trials. They are trained and tested on the HMBD database that contains 54,115 of the handwritten Arabic characters. Experiments are performed on the native CNN architectures and the best-reported testing accuracy is 91.96%. A transfer learning (TF) and genetic algorithm (GA) approach named “HMB-AHCR-DLGA” is suggested to optimize the training parameters and hyperparameters in the recognition phase. The pre-trained CNN models (VGG16, VGG19, and MobileNetV2) are used in the later approach. Five optimization experiments are performed and the best combinations are reported. The highest reported testing accuracy is 92.88%.